Cloud computing has one main Achilles heel in the form of increased latency. In most cases the cloud platform is not physically located near the data source, thus data transfer takes additional time. This can become highly problematic for the overall performance of latency sensitive applications and services.
The benefits of edge and fog computing are primarily reduced latency, coupled with the ability to extend the overall reach of the cloud service to the data source.
What is the Internet of Things (IoT)
Before we delve deeper into more detail of the various benefits of edge and fog computing, it’s worthwhile reviewing the subject of the Internet of Things (IoT) and the relationship with cloud services.
The Internet of Things (IoT) is the definition given to any electronic device that does not require human interaction and is able to connect to the Internet and share data with other connected devices.
These devices includes sensors, security cameras, smart speakers, smart fridges, etc. It is worth pointing out that laptops, PC’s, smartphones, tablets etc are technically not IoT devices, but rather ‘user devices’.
IoT is on the cusp of radically changing the technology landscape. Ericsson predicts that there will be 29 billion internet connected devices by 2022, and 18 billion of those will be related to IoT. Cisco have suggested that figure may be closer to 50 billion.
The amount of data expected to be in transit between IoT devices and the cloud is ever increasing. Our thirst for real-time analytics means unnecessary latency is a problem we can ill afford.
Is Cloud Computing Here For Good?
Having your data, virtual servers, storage and applications hosted in the cloud offers a huge number of business advantages. A recent article by Forbes has predicted that 83% of all enterprise workloads will be in the cloud by 2020. Thus it’s fair to surmise that cloud computing is here to stay considering it’s many benefits. There is a problem however.
The ‘fly in the ointment’ is our increasing demands on the cloud to provide services for real time applications and IoT devices. This is clearly not a match made in heaven when large distances are involved. Light moves pretty quick, but not always quick enough.
Overview of Fog Computing
Fog computing uses the concept of ‘fog nodes’ that reside either on the local LAN or a hop or two across the WAN of a private providers network. These fog nodes have higher processing and storage capabilities than edge IoT devices, and are located closer to the data source that with a certralised cloud computing solution.
Fog nodes can process the data from local edge IoT or user devices far quicker than sending the request to the cloud for centralised processing. This allows latency to be kept to a minimum for time sensitive applications and services. Data can also be sent by the fog nodes to the cloud for further centralised processing and storage if required.
For example, a user with a hand-held device wishes to review recent CCTV video stream from a locally located IoT security camera. As the camera does not have storage, the video stream will be requested from the cloud.
With fog computing, a local fog node can instead be responsible for the video stream, and is far quicker than offloading the processing to a centralised cloud platform.
Overview of Edge Computing
IoT and user devices are becoming increasingly powerful allowing more of your data to now be processed directly at the edge.
The edge computing model aims to ensure data is processed on the local IoT or user device itself rather than being sent to a fog node or all the way to the cloud for analysis. After being processed locally on the edge device, the data can still be sent to a fog node or the cloud for further intensive centralised processing and analysis.
By partially (or fully) processing the data on the edge, overall real-time performance is greatly enhanced. For example, a manufacturing system can have any number of IoT sensors capable of not only monitoring, but processing data locally to make adjustments to the manufacturing process in real-time.
Fog Computing Vs Edge Computing
Both fog computing and edge computing reduce latency by moving the compute element as close as possible to the data source to speed up processing of that data, although achieve this in slightly different ways.
Edge computing processes the data on the local IoT or user device, whereas fog computing allows the data to be processed on a more powerful local fog node located on the LAN or a hop or two across the WAN to a nearby datacentre.
Enhancing Cloud Computing
Edge and fog computing models compliment rather than replace cloud computing. Both design models ensure that time sensitive data can be processed locally either on the edge device or fog node without having to be sent back to the cloud. Any remaining relevant data can still be sent to the cloud for further analysis and storage.
10 Benefits of Edge and Fog Computing
1. Latency Reduction
Reduced latency is the primary benefit of edge and fog computing. Data does not necessarily need to be sent to the cloud for processing as the some of the compute can be performed nearer the data source for time sensitive services.
2. Improved Response Time
With a reduction in network latency, real-time applications will benefit from improved response time and greater overall user experience.
3. Enhanced Compliance
Data that can reside locally rather than moving to the cloud can increase compliance for certain business sectors.
4. Increased Security
Similar to compliance, if specific sensitive data does not need to move to the cloud for processing then overall security of that data will be increased.
5. Greater Data Privacy
Sensitive data can be processed locally and if required only a subset of that data be sent to the cloud for additional analytics.
6. Reduced Cost of Bandwidth
As certain data can be processed locally without being sent to the cloud, less network bandwidth will be required. With the ever increasing numbers of IoT devices all generating live data, this bandwidth saving could be considerable.
7. Overall Increase of Speed and Efficiency
If you have a number of local IoT and user devices that share data, allowing local processing between them rather than utilising cloud services will increase overall speed and efficiency of the service.
8. Less Reliance on WAN Services
Should overall access be lost to the internet or private cloud service due to a complete WAN failure, local services can continue to operate.
9. Greater Up-time of Critical Systems
Critical systems using the edge and fog computing model will have a greater uptime as the reliance on remote cloud services for data compute, analytics and storage is reduced.
10. Enhanced Services for Remote Locations
Systems running in remote or geographically challenging locations where access to the internet or private cloud services may be slow or unreliable will benefit from edge and fog computing.
Drawbacks of Fog and Edge Computing
Here are the three main disadvantages worth your consideration before you make the leap to fog or edge computing.
1. Increased Design Complexity
The use of more sophisticated edge IoT, user devices, and fog nodes on your network will increase complexity and overall support requirements.
2. Physical Security Considerations
Enhanced edge devices and fog nodes may be located in less secure environments than a central cloud platform in a secure data centre can provide.
3. Decentralised Design
The centralised model that cloud computing provides will make overall platform management and provisioning of hardware more time intensive.
Usage Examples For Specific Sectors
Both edge and fog computing design models are best suited for businesses that have a requirement for real-time data analysis and also perform instant action based on that data. Let’s take a look at 4 sectors that can benefit.
Internet of Things (IoT) is allowing innovative ways for brick-and-mortar stores to enhance overall customer experience. In-store staff can use handheld devices to provide customers with additional product information, check stock or perform on the spot payment transactions to reduce check-out queues.
Reducing network delay is crucial for these real-time customer services. Edge and fog computing are the perfect enhancement to the existing cloud computing model to deliver these services quickly and efficiently.
Edge and fog computing can greatly reduce the overall network delay for IoT devices responsible or collecting and analyzing real-time manufacturing data.
This can increase overall manufacturing efficiency as IoT sensors can now gather (and analyze) data locally in real-time. For example, the current condition of any part of the manufacturing process can now be automatically adjusted, refined and alerted on.
The benefits of edge and fog computing can be used to increase the performance of remote healthcare services by reducing latency. Wearable or implanted IoT devices can now be deployed to gather, monitor and analyse real-time patient data.
Patents are able to stay at home with their wearable IoT devices transmitting all required patient data directly to the hospital IT systems. This helps to reduce hospital waiting times and increase the overall patient experience.
Real-time data processing allows for data being collected across a city to be processed faster. Services such as traffic management to be greatly enhanced.
Traffic management systems can use edge and fog computing for real-time data analysis to alter traffic lights and intelligent road signs the moment an accident or road blockage occurs.
The benefits of edge and fog computing will ensure that a mixture of edge, fog and cloud computing will become the norm for most businesses. WiFi/LAN connected IoT and user devices will all be consuming a sites WAN bandwidth if their only option is to speak to the cloud for a particular service.
The reliance on cloud computing will continue to grow year on year. However the increasing amount of real-time data generated by IoT devices does not best suit to the centralised cloud design. The edge and fog computing models address this issue and will encourage a shift to a more hybrid design.
Both edge and fog computing offer a number of advantages in a business world. This is becoming more reliant on real-time analytics data to keep competitive.
Please get in touch to discuss your edge and fog networking requirements in more detail